持续创作，加速成长！这是我参与「掘金日新计划 · 10 月更文挑战」的第21天，点击查看活动详情
4.1 write（）函数
import pyautogui
# 使用write（）只能按单字符键，因此不能按Shift 或 F1键
# 立即输入字符
pyautogui.write('Hello world!')
# 依次输入字符，每个字符之间间隔0.25秒
pyautogui.write('Hello world!', interval=0.25)
4.2 press（）、keyDown（）和keyUp（）函数
import pyautogui
# press（）函数实际上只是keyDown（）和keyUp（）函数的包装器，它们模拟按下一个键，然后释放它。
# 在每次按下之间添加延迟间隔，设置interval即可，传入int或float
# 按下回车键
pyautogui.press('enter')
# 按下F1键
pyautogui.press('f1')
# 按下←键
pyautogui.press('left')
# 要在按住Shift键的同时按左箭头键三次
pyautogui.keyUp('shift')
pyautogui.press(['left', 'left', 'left']) # 多个按键用字符串列表
pyautogui.keyDown('shift')
# 设置presses=3，实现按←键3次
pyautogui.press('left', presses=3)
4.3 hotkey() ：实现热键组合
import pyautogui
# 为了方便按下热键或键盘快捷键，可将热键（）传递给几个键串，这些键串将按顺序按下，然后按相反顺序释放。
# 要在每次按下之间添加延迟间隔，可为interval参数传递int或float
pyautogui.hotkey('ctrl', 'shift', 'esc')
# 等价于
pyautogui.keyDown('ctrl')
pyautogui.keyDown('shift')
pyautogui.keyDown('esc')
pyautogui.keyUp('esc')
pyautogui.keyUp('shift')
pyautogui.keyUp('ctrl')
4.4 KEYBOARD_KEYS
KEY_NAMES = [
"\t",
"\n",
"\r",
" ",
"!",
'"',
"#",
"$",
"%",
"&",
"'",
"(",
")",
"*",
"+",
",",
"-",
".",
"/",
"0",
"1",
"2",
"3",
"4",
"5",
"6",
"7",
"8",
"9",
":",
";",
"<",
"=",
">",
"?",
"@",
"[",
"\\",
"]",
"^",
"_",
"`",
"a",
"b",
"c",
"d",
"e",
"f",
"g",
"h",
"i",
"j",
"k",
"l",
"m",
"n",
"o",
"p",
"q",
"r",
"s",
"t",
"u",
"v",
"w",
"x",
"y",
"z",
"{",
"|",
"}",
"~",
"accept",
"add",
"alt",
"altleft",
"altright",
"apps",
"backspace",
"browserback",
"browserfavorites",
"browserforward",
"browserhome",
"browserrefresh",
"browsersearch",
"browserstop",
"capslock",
"clear",
"convert",
"ctrl",
"ctrlleft",
"ctrlright",
"decimal",
"del",
"delete",
"divide",
"down",
"end",
"enter",
"esc",
"escape",
"execute",
"f1",
"f10",
"f11",
"f12",
"f13",
"f14",
"f15",
"f16",
"f17",
"f18",
"f19",
"f2",
"f20",
"f21",
"f22",
"f23",
"f24",
"f3",
"f4",
"f5",
"f6",
"f7",
"f8",
"f9",
"final",
"fn",
"hanguel",
"hangul",
"hanja",
"help",
"home",
"insert",
"junja",
"kana",
"kanji",
"launchapp1",
"launchapp2",
"launchmail",
"launchmediaselect",
"left",
"modechange",
"multiply",
"nexttrack",
"nonconvert",
"num0",
"num1",
"num2",
"num3",
"num4",
"num5",
"num6",
"num7",
"num8",
"num9",
"numlock",
"pagedown",
"pageup",
"pause",
"pgdn",
"pgup",
"playpause",
"prevtrack",
"print",
"printscreen",
"prntscrn",
"prtsc",
"prtscr",
"return",
"right",
"scrolllock",
"select",
"separator",
"shift",
"shiftleft",
"shiftright",
"sleep",
"space",
"stop",
"subtract",
"tab",
"up",
"volumedown",
"volumemute",
"volumeup",
"win",
"winleft",
"winright",
"yen",
"command",
"option",
"optionleft",
"optionright",
]
PyAutoGUI利用PyMsgBox中的消息框函数提供跨平台的纯Python方式来显示JavaScript样式的消息框。
5.1 alert（）
import pyautogui
pyautogui.alert(text='这里写内容', title='通知框标题位置', button='OK')
5.2 confirm()
import pyautogui
pyautogui.alert(text='这里写内容', title='通知框标题位置', button='OK')
pyautogui.confirm(text='这里写内容', title='确认框标题位置', buttons=['OK', 'Cancel'])
pyautogui.prompt(text='这里写内容', title='输入标题位置' , default='默认输入')
5.3 prompt()
import pyautogui
pyautogui.prompt(text='这里写内容', title='输入标题位置' , default='默认输入')
5.4 password()
import pyautogui
pyautogui.password(text='这里写内容', title='密码框', default='', mask='*')
6 屏幕截图与图像定位
PyAutoGUI可以拍摄屏幕截图，将其保存到文件中，并在屏幕中定位图像。OSX使用操作系统附带的screencapture命令。Linux使用scrot命令，可以通过运行sudo-apt-get-install-scrot来安装该命令。
功能介绍：一个需要点击的按钮，并且想在屏幕上找到它。
6.1 屏幕截图
import pyautogui
# 截取全屏 在1920 x 1080屏幕上，screenshot（）函数大约需要100毫秒-不快但不慢。
im1 = pyautogui.screenshot()
# 截取全屏，并以图片保存
im2 = pyautogui.screenshot('my_screenshot.png')
# 截取指定位置，传递要捕获的区域的左侧、顶部、宽度和高度的四个整数元组：
im = pyautogui.screenshot(region=(0,0, 300, 400))
6.2 定位单个目标
import pyautogui
# ---------------------------------------------------------------
# 获取感兴趣区域的 (left, top, width, height)
button7location = pyautogui.locateOnScreen('looksLikeThis.png')
print(button7location)
print(button7location[0])
print(button7location.left)
# 计算感兴趣区域的中心点的xy坐标
button7point = pyautogui.center(button7location)
print(button7point)
print(button7point[0])
print(button7point.x)
# 点击感兴趣区域的中心点坐标
button7x, button7y = button7point
pyautogui.click(button7x, button7y)
# ---------------------------------------------------------------
# 快速点击感兴趣区域
pyautogui.click('looksLikeThis.png')
# ---------------------------------------------------------------
# 设置置信度 需要安装opencv
button7location = pyautogui.locateOnScreen('looksLikeThis.png', confidence=0.9)
print(button7location)
print(button7location[0])
print(button7location.left)
# ---------------------------------------------------------------
# 获取感兴趣区域的中心点位置坐标，并且点击
x, y = pyautogui.locateCenterOnScreen('looksLikeThis.png')
pyautogui.click(x, y)
6.3 定位全部目标的位置
import pyautogui
# 这些“定位”功能相当昂贵；他们可以用整整一秒钟的时间跑。
for pos in pyautogui.locateAllOnScreen('someButton.png')
print(pos)
# 提高速度的最佳方法：传递一个区域参数（一个4整型元组（左、上、宽、高）），以仅搜索屏幕的较小区域，而不是全屏：
pyautogui.locateOnScreen('someButton.png', region=(0,0, 300, 400))
6.4 灰度匹配
import pyautogui
button7location = pyautogui.locateOnScreen('looksLikeThis.png', grayscale=True)
print(button7location)
6.5 像素匹配（获取屏幕截图中像素的RGB颜色）
import pyautogui
# 获取屏幕截图中像素的RGB颜色方案①
im = pyautogui.screenshot()
color_RGB = im.getpixel((100, 200))
print(color_RGB) #(130, 135, 144)
# 获取屏幕截图中像素的RGB颜色方案②
pix = pyautogui.pixel(100, 200)
print(pix) # RGB(red=130, green=135, blue=144)
print(pix[0]) # 130
print(pix.red) # 130
# 如果只需要验证单个像素是否与给定像素匹配，请调用pixelMatchesColor（）函数，将其表示的颜色的X坐标、Y坐标和RGB元组传递给它：
pyautogui.pixelMatchesColor(100, 200, (130, 135, 144)) # True
pyautogui.pixelMatchesColor(100, 200, (0, 0, 0)) # False
# tolerance关键字参数指定红色、绿色和蓝色值在仍匹配时可以变化多少：
pyautogui.pixelMatchesColor(100, 200, (140, 125, 134)) # False
pyautogui.pixelMatchesColor(100, 200, (140, 125, 134), tolerance=10) # True